Red Hat Bugzilla – Bug 77386
Assertion failure in journal_write_metadata_buffer() at journal.c:408: "buffer_jdirty(jh2bh(jh_in))"
Last modified: 2005-10-31 17:00:50 EST
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 5.0)
Description of problem:
My server running SMP 220.127.116.11 kernel (RedHAt 7.3).
Equipeded with 2 P3 933MHz and 1 GB RAM
Suddenly hanged up on me. Still pingable but all services dead. Upon rebooting,
look through the log file and found the kernel error code. I've attached as
Any idea why is that so?
Version-Release number of selected component (if applicable):
I search thru the repository and can only find a similar entry titled
Assertion failure in journal_write_metadata_buffer() at
But what i got at my system is c:408 instead of c:406. Not too sure whats the
Created attachment 83806 [details]
LogFile Error code
Have you seen any similar problems before? Was there _anything_ else in the
logs related to possible ext3 or IO problems?
This is the 2nd time I encounter it on the same machine in 4 months time. Not
very frequent, but is interested to know whats the cause of it
Created attachment 89554 [details]
Error message as found in the system logfile after rebooting.
I posted the logfile yesterday, here is the explination: We recieved this error
on our server running Red Hat 7.3 and 2.4.18-5smp. The server is a dedicated
postgresql database server. I noticed in the last message for bug # 64107 that
the error was recieved on heavy IO load, and I think this might have been a
similar problem for us. We had turned on query logging on our database which
was producing data at a rate of about 2.6GB in around 6 hours. Once the error
occured, we had no access to the server and had to power off/on.
2.4.18-3 had a problem that triggered this quite frequently, and that was fixed
in 2.4.18-5. However, I found another rare case that could sometimes still
cause it even in that kernel, and that should be fixed in all subsequent errata
later than 2.4.18-5. Please reopen this report if you still have the problem on
a current kernel.